TEXT
COMMERCIAL
CAPTIONED
Paste an early draft of something you wrote and the version you sent, then explain what changed and why it is better. Typo-only differences, missing explanations, fabricated bad drafts and identical texts are refused; one pair per contributor, from a global base strongest in Southeast Asia, the United States and Europe, so English runs from first- to second-language writers and edits cover register, clarity and tone as much as grammar. Drafts get deleted, so human writing's intermediate state is almost never preserved, and the existing error-correction corpora (W&I+LOCNESS, JFLEG) are non-commercial essay sets. The explanation, written by the only person who knows why, turns a revision pair into supervision for preference, rewriting and feedback models.
Volume
5,000+
Pairs
Contributors
5,000+
Regions
100+ countries
Updated
What’s in the collection and how it’s delivered.
Data modality
Text
Annotated
Captioned (author's edit explanation)
Exclusivity
Non-exclusive rights
Capture tool
Typed in the Kled app
Use cases
Preference and rewriting model supervision, writing feedback generation, revision intent modeling
Format
Plain text
Resolution
Plain text, UTF-8
Total volume
5,000+ pairs (approx. 25 MB)
Regions
100+ countries
Purity level
99%+ passed QA review
Where this data comes from and how it’s licensed.
Every file comes straight from the contributor who provided it and is rights-cleared before it’s listed, so your license traces back to the source.
Contributor
Provides the data and uploads it straight to Kled.
Kled
Clears the rights before anything is listed.
Your team
Licenses it under the terms below.
License terms
Non-exclusive commercial training license.
Consent
Each contributor opted in to this use when they uploaded.
Specs reflect the current collection and may change as new data is added. Final terms are set in your license agreement.
